To address the issue of outlier removal in remote sensing image feature matching, a novel algorithm, local-angular-length coherence constraint (LALC), based on geometric is proposed. This approach leverages positional relationship between adjacent inliers to reformulate problem as identifying consistent structures. First, by using local constraint, K-nearest neighbours (KNN) structure can be constructed, and vectors obtained. Subsequently, analysing across images, both angular length...

Since the day it was proposed, return-oriented programming has shown to be an effective and powerful attack technique against write or execute only (W ⊕ X) protection. However, a general belief in previous research is, systems deployed with address space randomization where executables are also randomized at run-time able defend programming, as addresses of all instructions randomized. In this paper, we show that due weakness current technique, there still ways launching attacks those...

Android Application Framework is an integral and foundational part of the system. Each 1.4 billion devices relies on system services to manage applications resources. Given its critical role, a vulnerability in framework can be exploited launch large-scale cyber attacks cause severe harms user security privacy. Recently, many vulnerabilities were exposed, showing that it vulnerable exploitable. However, most existing research has been limited analyzing applications, while there are very few...

Nowadays, deep learning has been widely used. In natural language learning, the analysis of complex semantics achieved because its high degree flexibility. The deceptive opinions detection is an important application area in model, and related mechanisms have given attention researched. On-line are quite short, varied types content. order to effectively identify opinions, we need comprehensively study characteristics explore novel besides textual emotional polarity that used text analysis....

Artificial Intelligence (AI) frees humans from the heavy calculation burden in science and engineering.The AI can analyze mass data on classroom behavior backend of intelligent teaching platforms, extract useful features student behaviors class.Based feature fusion, this paper develops a novel method to class participation, which correlates participation with behaviors.Firstly, basic for analysis were processed, including records online video, relevant formulas constructed.After that,...

Purpose The purpose of this paper is to summarize the existing point cloud target detection algorithms based on deep learning, and provide reference for researchers in related fields. In recent years, with its outstanding performance 2D images, learning technology has been applied light ranging (LiDAR) data improve automation intelligence level detection. However, there are still some difficulties room improvement from 3D cloud. paper, vehicle LiDAR method chosen as research subject....

As an advanced component of embedded systems, SoC issues excellent performance in size, reliability and operating efficiency. Its application is popular. mainly composed some computing cores IP modules. design concerned to cooperation hardware HW software SW. In this paper, a method HW-SW co-design for discussed. example, the display modules issued. The solution based on FPGA with VHDL software. Simulation test show that function replaced by hardware. With new modules, can be written more...

Controlled vocabularies have been used as the means for unifying disparate terminologies found within an application field. This unification leads to better administration of information and enhanced communication among various parties. Semantic networks shown be excellent vehicles modeling controlled vocabularies. However, they often lack necessary access flexibility robustness required by external agents such intelligent information-locators decision-support systems. In this paper, we...

Radio frequency (RF) stealth is one of the essential research hotspots in radar field. The anti-sorting signal an important direction RF signal. Theoretically speaking, design based on failure principle sorting algorithm, and SDIF algorithm a core widely used engineering. Thus, this paper, first analyzed detail. It pointed out that threshold function will fail when pulse repetition interval (PRI) value obeys distribution whose length 20 times larger than minimum PRI. Secondly, correctness...

Radio frequency (RF) stealth anti-sorting technology is a research hotspot in the radar field. In this study, signal design principles of anti-cluster and anti-SDIF sorting were investigated for processes clustering pre-sorting sequence-difference-histogram main sorting. Then, accordance with principle, 2D interleaving feedback hyperchaotic system based on cosine-exponential was designed. A method to modulate pulse repetition interval (PRI) parameters carrier wide intervals through...

Radio frequency (RF) stealth anti-sorting technology can improve the battlefield survival rate of radar and is one research hotspots in field. In this study, signal design principle anti-sequential difference histogram (SDIF) sorting was explored for main algorithm SDIF. Furthermore, we designed a piecewise linear chaotic system with interval number parameterization based on random disturbance proposed method to modulate repetition period widely spaced pulses using system. Then, considering...

Cluster ensemble is an effective method to improve the effect in data clustering, but results of existing cluster algorithms are usually not so good when they process mixed attributes datas, main reason that still dispersed. To solve this problem, paper presents a new weighted algorithm based on graph theory. It first clusters datasets and gets members, then sets weights each object with proposed function, determines relationship between data-pair by setting edges them, it can get nearest...

Controlled vocabularies (CVs) are networks of concepts that unify disparate terminologies and facilitate the process information sharing within an application domain. We describe a general methodology for representing existing CV as object-oriented database (OODB), called vocabulary repository (OOVR). A formal description OOVR methodology, which is based on structural abstraction technique, given, along with algorithmic number theorems pertaining to some methodology's characteristics. An...

China has been one of the countries with biggest manufacturing industry in world. An advanced or a world-class is pursued target for Chinese industries. can be set up through modern automation systems. And system built multi-layer network structure, which includes ERP, MES and FCS. All three layers are depended on embedded systems technology practice. This paper to focus analyzing relationship between systems, discussing influence their networks development an manufacturing, especially China.

Randomness and fuzziness are two fundamental kinds of uncertain phenomena, which can be handled respectively by probability theory (sigma system) possibility (max system). For many practical problems information processing, random uncertainty may often co-exist with fuzzy uncertainty, reminds us to look for a novel mechanism called sigma-max hybrid inference. This cope randomness jointly achieve direct fusion heterogeneous modeled or possibility. Specifically, we interested in typical forms...

Recently, multiantenna spectrum sensing (MSS) using de-ep neural network has received much attention. However, most of them are not explicitly formulated to account for non-circular sources. In practice, the primary users' signals may come from either circular or noncircular sources, both. We are, therefore, motivated investigate MSS problem hybrid and source in cognitive radio. Specifically, we first introduce augmented sample covariance matrix (ASCM) exploit additional statistical...

In view of the slowness and locality convergence for basic genetic algorithm (BGA short) in solving complex optimization problems, we proposed an improved named self-adpative multi-population composite (SM-CGA short), gave structure implementation steps algorithm; then consider its global under elitist preserving strategy using Markov chain theory, analyze performance through three examples from different aspects. All results indicate that new possess interesting advantages such as better...

Light detection and ranging (LiDAR) point clouds are sparse, unstructured, disordered; hence, traditional convolutional neural networks unsuitable for direct application in point-cloud data processing. Graph convolution (GCNs) can be used to process having the aforementioned characteristics; however, they inefficient when adjacent relationship of cloud is uncertain adjacency-matrix elements abundant. To solve these problems, a ground filtering method based on superpoint graph network (SPGCN)...

Motion estimation is the most important step in video compression. Most of current compression systems use forward motion estimation, where information derived at encoder and sent to decoder over channel. Backward does not derive an explicit representation encoder. Instead, implicitly embeds alternative subspace. recently, algorithm that adopts least-square prediction (LSP) for backward has shown great potential further improve coding efficiency. Forward have both their advantages...

This paper managed to induce probability theory (sigma system) and possibility (max respectively from randomness fuzziness, through which the premature of is expected be well founded. Such an objective achieved by addressing three open key issues: a) lack clear mathematical definitions fuzziness; b) intuitive definition possibility; c) abstraction procedure axiomatic probability/possibility their definitions. Especially, last issue involves question why axiom "maxitivity" adopted for...

Clustering ensemble selection has been confirmed that it can always achieve better result than traditional clustering algorithms. However, many selective algorithms cannot eliminate the inferior quality partitions' influence and accuracy of is not high. In order to solve problems, paper proposes a new algorithm. The algorithm, firstly, uses validity evaluation evaluate all available partitions selects best as reference partition; secondly, defines strategy via diversity; lastly, adaptive...